一键在 Manus 中运行任何 Skill

开始使用

systematic-debugging

星标3

分支2

更新时间2026年4月8日 17:07

Systematically diagnose and fix bugs using logs, stack traces, and targeted investigation.

安装

用 Codex 或 Claude 帮你安装复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。

在 Manus 中运行

来源

dbczumar

dbczumar/agent-plane

打开 GitHub 仓库查看创作者相关仓库

下载

在 Manus 中运行

相关职业SOC

基于 SOC 职业分类

软件开发工程师计算机与数学类职业·SOC 15-1252

SKILL.md

readonly

name	systematic-debugging
description	Systematically diagnose and fix bugs using logs, stack traces, and targeted investigation.

When debugging:

Reproduce first: Run the failing case and capture the exact error. Don't guess — observe.
Read the stack trace bottom-up: The root cause is usually near the bottom. Work upward to understand the call chain.
Form a hypothesis, then test it: Don't shotgun-debug. Identify one likely cause, verify it, then move on if wrong.
Check the boundaries: Most bugs live at integration points — API calls, file I/O, type conversions, serialization. Check inputs and outputs at each boundary.
Verify the fix: Run the original failing case again. Then check that you didn't break the happy path.

Approach:

Start with grep or glob to find the relevant code
Read the error logs or stack trace carefully
Add targeted debug output if needed (remove it after)
Fix the root cause, not the symptom
Run tests to confirm

同仓库更多 Skills

同仓库

agent-plane-knowledge

dbczumar/agent-plane

Deep reference on agent plane config format, executor types, skill/tool structure, and conventions. Load when you need to look up how the platform works.

2026-04-193

generate-agent

dbczumar/agent-plane

Patterns and templates for generating valid agent plane agent directories. Load when ready to create files.

2026-04-193

repl-sdk

dbczumar/agent-plane

Build terminal REPLs for agent-plane using the Python UI SDK. Use when the user asks about building a REPL, terminal chat, or CLI frontend for an agent.

2026-04-193

deploy-databricks-app

dbczumar/agent-plane

Deploy agent-plane to Databricks Apps with Lakebase and UC Volumes. Use when setting up infrastructure, deploying, redeploying, troubleshooting, or connecting to a remote agent-plane instance on Databricks.

2026-04-153

detect-framework

dbczumar/agent-plane

Detect Python agent frameworks from code imports and map them to agent plane executor types. Load when the user has existing agent code to integrate.

2026-04-123

deep-research

dbczumar/agent-plane

Investigate a topic in depth, find sources, and synthesize findings.

2026-04-103

name	systematic-debugging
description	Systematically diagnose and fix bugs using logs, stack traces, and targeted investigation.

When debugging:

Reproduce first: Run the failing case and capture the exact error. Don't guess — observe.
Read the stack trace bottom-up: The root cause is usually near the bottom. Work upward to understand the call chain.
Form a hypothesis, then test it: Don't shotgun-debug. Identify one likely cause, verify it, then move on if wrong.
Check the boundaries: Most bugs live at integration points — API calls, file I/O, type conversions, serialization. Check inputs and outputs at each boundary.
Verify the fix: Run the original failing case again. Then check that you didn't break the happy path.

Approach:

Start with grep or glob to find the relevant code
Read the error logs or stack trace carefully
Add targeted debug output if needed (remove it after)
Fix the root cause, not the symptom
Run tests to confirm